Sunday, 20 July 2014
Brentano on Mental Acts & Mental Objects
Franz Brentano offers us a view of mental representation that is an example of self-consciousness.
We don't just ‘have a representation’, or ‘see one’, or ‘become aware’ of one, we are also at the same time aware of the representing. Prima facie, it seems strange that Brentano argued that we ‘represented’ a representation to ourselves, rather than simply had one. The way he puts the representing of representations is that it's a cognitive or volitional process. It's often thought to be the case, say, with sense impressions and sense-data that they are just ‘given’ to us without cognitive activity. That is why the term ‘the given’ was so important to 20th century empiricist epistemology. That very term stressed the supposed fact that sense-data aren't the result of any cognitive activity and therefore must be unpolluted by our prior concepts or categories.
For instance, when I look at my white fridge, seemingly non-cognitively, white, square (not literally square!), and other kinds of sense-data are just given to me – I simply have them or, in a way, find them in my mind [see Ned Block’s 1992/1997]. All this seems to distinguish sense-data from Brentano’s ‘representation’ because he argued that we cognitively ‘represent’ the representation to ourselves.
Of course we must first ask what it actually means to ‘represent a representation’. Can a representation also represent something or other without our own cognitive representing of the representation that itself represents something external? ‘Representing’ implies that we make some kind of cognitive decision to become aware of a representation that would otherwise be un-cognised (even if an un-cognised representation makes no sense!).
Take the representation of a particular sound. When we have a sound-representation, we don’t just have it, or become aware or conscious of it, there must be a secondary cognitive process that becomes part of the sound-representation and all conscious representations. We must also be conscious of ‘the act of hearing it’. In that case, the sound-representation must come along with what must be another representation or awareness – that of the act of awareness itself. So we now have:
i) the sound-representation
ii) the act-representation (of being conscious of the sound-representation).
In that case, all representations must have binary components. We aren't only conscious of the representation itself: we must also be conscious of the act of representing that representation to ourselves. We can say that the act-representing is of a higher order. And the representation-object of the act is of a lower order. Though, in all genuine representations, that higher-order act must always accompany each conscious representation.
Brentano now stresses the fact that although representations are of this binary nature, each conscious representation doesn't also require two acts. The representation itself is seen as the ‘first object’ by Brentano. And the mental act of representing, or being conscious of representing, is the ‘second object’. The second object, or mental act, can be deemed as ‘reflexive’ in that it's a mental process or act which has its own object, which itself becomes the object of a reflexive act of awareness about the mental act itself. In that case, couldn’t we also become aware not only of the mental act with a representation as its object, but an act which has another mental act, rather than a representation, as its object? This third-level process would include two acts of cognition, where one mental act itself becomes the mental object of a third-level mental act. This process, clearly, could continue indefinitely. However, despite that possible higher level indefinite regress, Brentano himself is aware of the possible lower-level ‘Cartesian presumption’ that would lead to an endless regress of mental acts. This, however, is about the lower-level relation between a conscious act of awareness and a representation not to be deemed a mental act in itself; but only a mental object of various cognitive acts. That is, if the initial representation were itself an act, though still a mental object of a higher mental act, then that representation-act, or mental object-act, would itself have its own mental object or representation. We would have to move one step further beyond what we took to be the pure representation and non-cognitive starting point to its own representation. But if we go further than the initial pure representation, then we could also see the initial representation’s own representation-object as also being a representation-act as well. This process, clearly, could go on indefinitely, just as my higher-level version could.
So the ‘Cartesian presumption’ must be that we must reach the foundation or bedrock of such mental processes at some time. Therefore some representations, or Cartesian ‘ideas’, must be pure representations and not at all representation-acts.
This train of thought is analogous to Wittgenstein’s position in his Tractatus that there must be ‘logically proper names’ that are no longer ‘analysable’ and must therefore be basic and fundamental. (These names too must be seen to halt a similar indefinite regress, just as his ‘simple objects’, the objects of the ‘simple names’, called a halt to a possible ontological and logical indefinite regress.)
However, if we call a halt to the Cartesian indefinite regress by accepting the existence of a pure representation or representation-object, then what of our higher-level case of an indefinite regress of mental acts, or higher-order acts which take lower-order acts as mental objects; though which are in fact still act-objects and never pure representation-objects. In the higher-order case we can have an act of an act of an act… And so on. Just as in the Cartesian case we could have a mental act of a mental act… All this is instead of an act which has as its object a pure representation that is in no way itself a mental act. If we can call a halt by positing a lower-order representation, perhaps in our higher-order version we could take a specific higher-order mental act as the end of a short regress, as it were. The lower-order representation would then therefore be foundational and fundamental, whereas our higher-order mental act would be the end-point or final consequence of a small foundational structure supported by our pure representation.
In the case of the earlier sound-representation, we can take our higher-order approach of taking a possible indefinite regress of mental acts rather than the lower-order Cartesian approach of the indefinite regress of trying to find a representation that isn't also a mental act which has its own object-representation. In terms of being aware of an act rather then of a representation, it would mean that
to be aware of a sound is to be aware of being aware of a sound, and then to be aware of being aware of a sound would involve being aware of that awareness, and so on indefinitely. We can now ask: How do we, or how can we, call a halt to this potential indefinite regress? Would we need to treat a higher-level act in roughly the same way as we treated our foundational representation? Instead of seeing a higher-order mental act as foundational, we could see it as an end-point or some kind of cognitive brick wall which we couldn’t break through.
Brentano himself calls a halt to this potential regress of acts in this way.
Firstly, he rejects the higher-order act of being aware of our awareness of a sound. This means, quite simply, that awareness itself can't be a mental object of another mental act of awareness. In other words, all we really have is a second-order acts and first-order, or foundational, representations. We must stick with this binary relation and not believe that there are also tertiary or yet higher-order cognitive acts which take other cognitive acts as their mental objects. However, at a prima facie level, it does seem possible to be aware of an instance of our awareness, or to take a cognitive act as an object-act, rather than as a pure object or representation.
How does Brentano defend his sudden halt to higher-order acts of awareness?
Brentano argues that this ostensible case of an act of awareness which takes another act of awareness as its cognitive object is in fact a different kind of a mental act from the lower-order act of simply being aware of a sound. So, according to Brentano, the ostensible act of being aware of being aware of a sound is no different to the more basic being aware of a sound. At a prima facie level, again, the third-order awareness of an act of awareness seems possible. How does Brentano actually defend his rejection of third-order acts of awareness?
If we were to ‘observe a mental act’ it would be an attempt to make it an example of Brentano’s ‘first object’. In that case, we would find ourselves with two first objects – the initial pure representation and an act of awareness of that representation which is now itself a ‘first object’ of another mental act of awareness. Clearly, we can't have two first objects. Brentano therefore concludes that an act, rather than a pure representation, could never be taken as a first object but must always be taken as a ‘second object’.
Even though we are talking here about mental acts and mental objects of those acts, Brentano rather strangely brings in the notion of ‘observation’. He makes the point that when we talk of observation we presume a distinction between observer and observed. Clearly, in that case of a pure representation and its act, the representation is analogous to the observed and the act of awareness of that representation is analogous to the observer. What would be an act of awareness of another object? We would therefore have two acts or processes of awareness occurring at the same time, each with its own mental object. If we think in terms of two cognitions or two sub-vocal expressions of two different propositional beliefs occurring at the same time, such scenarios seem even more unlikely. Take two examples:
i) Tony Blair is a liar.
ii) Venus is a planet.
Such sub-vocal expressions could not occur at the same time in the same subject. This becomes even clearer if we say that the various objects, i.e., Tony Blair, Venus, planets, liars, etc. must somehow be cognised at more or less the same time. Not only that: two different mental acts of predication must occur at the same time. We couldn't observe a mental act that is just as much an act or process as the mental observation itself. In this case, an act of observation and an act of awareness must occur at the same time. So an act can't be a first object because it is not in fact a genuine mental object at all – it is a mental act or process. An act or process cannot be the mental object of a further mental act because one act would need to occur at the same time as another mental act.
So we were wrong to see mental acts as becoming the possible mental objects of further mental acts. We assumed that an act or process could itself become the mental object of another mental act. We failed to realise that an act, even as seen as a mental object, must still be a mental act in the scenario of an act of awareness that takes another act of awareness as its mental object or representation. According to the first case, the representation was the observed and the act was the observer. But now we have two acts, even when one act is taken to be a mental object-act. If we have two acts, then we must have two observers and no genuine example of a pure observed representation. Prima facie, the idea of an observer observing an observer seems odd. However, it does actually seem possible, as in the case of a spy of one country spying on a spy of another country. In that scenario, why can’t a mental act take another mental act as its mental object, or even as its quasi-object? Again, an observer can take another observer as his object of observation. So just as an observer can take on two roles, as observer and observed, why can’t a mental act also be a mental object for a further higher-order mental act? However, two cognitive mental acts cannot occur simultaneously. In that case, it's simply incorrect to see a mental act as also a potential mental object of another mental act. Unlike a representation, a mental act is a kind of mental process. A mental process can take a representation as its mental object because the representation is not itself a process. However, in order for a mental act also to become a mental object-act, we would need to accept two simultaneous and different mental processes occurring at the same time in the same mind. Clearly this couldn’t happen. For example, we would need to become aware of a mental act which must also, as a mental process, itself be in the mind if taken as the object-act of a further mental act. It should have been clear that this higher-order mental process would require a mental act, better seen as a temporal mental process, occurring at the same time as a mental act or temporal process in the mind which takes it as its mental object. But that higher-order mental act couldn't take an act as a mental object if that act itself is a temporal process of awareness of a lower-order mental object. And if this higher-order mental act can't actually be a genuine first object, then the yet higher-order mental act we believed took such a lower-order mental act as its mental object couldn't itself be a genuine higher-order mental act simply because it doesn't actually have its own higher-order mental object-act. Such ostensible higher-order mental processes and mental objects would therefore be like a single person playing the piano at the same time as doing a crossword or drinking a cup of tea and a can of lager concurrently. Looking for such higher-order mental objects and mental acts is also like a dog trying to catch its own tail or trying to look at our own eyes without the benefit of a mirror or reflector.
Conclusion: Brentano and Introspection
Finally, there may be a problem with Brentano’s use of the word ‘observation’ to get his arguments across. It depends, however, if he really believed that mental acts effectively - or literally - observe representations, or whether it's just an analogy on his part. Clearly, ‘observation’ is a term connected to the faculty of vision, which couldn't be applicable to any mental process or mental act. If Ryle, amongst others, rejected ‘the eye of the mind’ [Ryle, 1949], or the ‘internal theatre’ [Dennet, 1991], or introspective acts of perception of the mental events and mental objects on the mind’s stage [Davidson, 1980], so we should be equally suspicious of Brentano’s use of the term ‘observation’.
If it's not the case that mental acts observe mental representations, then how, in fact, can we make sense of the relation between an act of awareness and its representation-object, or of any other such introspective processes in the mind for that matter [see also Wittgenstein, 1953/57]? If we deny such quasi-visual metaphors, if they are metaphors in Brentano’s mind, then we may have to give up the notion of introspection itself, and even, as Ryle and Churchland do, the very ‘concept of mind’ as it is - or has been - seen traditionally.
Notes and Further Reading
Block, N – (1992/1997) ‘Begging the Question Against Phenomenal Consciousness’, in The Nature of Consciousness, ed. N. Block, A Bradford Book
Brentano, F – (1874) Psychology from an Empirical Standpoint
- (1911) On the Classification of Psychical Phenomena
Davidson, D – (1980) ‘Knowing One’s Own Mind’, from his Essays on Actions and Events, Oxford University Press
Dennet, D – (1991) ‘The Cartesian Theatre’, in Consciousness Explained, Boston: Little, Brown, and Co.
- (1981/1997) ‘Skinner Skinned’, in his Brainstorms, Penguin Books
Ryle, G – (1949) The Concept of Mind, Penguin Books
Wittgenstein, L – (1953/57) Philosophical Investigations